OSDN Git Service

android-x86/external-parted.git
16 years ago* maint.mk: Merge from coreutils.
Jim Meyering [Thu, 5 Jun 2008 07:10:57 +0000 (09:10 +0200)]
* maint.mk: Merge from coreutils.

16 years agogpt_write: fix a write-uninitialized error
Jim Meyering [Fri, 30 May 2008 19:10:06 +0000 (21:10 +0200)]
gpt_write: fix a write-uninitialized error

* libparted/labels/gpt.c (gpt_partition_new):
  ==32570== Syscall param write(buf) points to uninitialised byte(s)
  ==32570==    at 0x36CF8D6540: __write_nocancel (in /lib64/libc-2.8.so)
  ==32570==    by 0x443D29: linux_write (linux.c:1642)
  ==32570==    by 0x414E4E: ped_device_write (device.c:370)
  ==32570==    by 0x4404CE: gpt_write (gpt.c:1091)
  ==32570==    by 0x418F16: ped_disk_commit_to_dev (disk.c:486)
  ==32570==    by 0x418F5C: ped_disk_commit (disk.c:509)
  ==32570==    by 0x40BAD6: do_mkpartfs (parted.c:981)
  ==32570==    by 0x40A035: command_run (command.c:139)
  ==32570==    by 0x412150: non_interactive_mode (ui.c:1540)
  ==32570==    by 0x40EE83: main (parted.c:2487)
  ==32570==  Address 0x501b63a is 58 bytes inside a block of size 16,384 alloc'd
  ==32570==    at 0x4A04FC0: memalign (vg_replace_malloc.c:460)
  ==32570==    by 0x4A0507A: posix_memalign (vg_replace_malloc.c:569)
  ==32570==    by 0x443CED: linux_write (linux.c:1637)
  ==32570==    by 0x414E4E: ped_device_write (device.c:370)
  ==32570==    by 0x4404CE: gpt_write (gpt.c:1091)
  ==32570==    by 0x418F16: ped_disk_commit_to_dev (disk.c:486)
  ==32570==    by 0x418F5C: ped_disk_commit (disk.c:509)
  ==32570==    by 0x40BAD6: do_mkpartfs (parted.c:981)
  ==32570==    by 0x40A035: command_run (command.c:139)
  ==32570==    by 0x412150: non_interactive_mode (ui.c:1540)
  ==32570==    by 0x40EE83: main (parted.c:2487)

  reproduce with this:
  dev=f
  dd if=/dev/null of=$dev bs=1 seek=30M 2>/dev/null
  ./parted -s $dev mklabel gpt
  valgrind ./parted -s $dev mkpartfs primary ext2 0 16795000B

16 years agoplug leaks in do_print
Jim Meyering [Thu, 29 May 2008 20:35:06 +0000 (22:35 +0200)]
plug leaks in do_print

* parted/parted.c (do_print):
25 bytes in 8 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416E67: ped_malloc (libparted.c:234)
   by 0x41751F: ped_strdup (unit.c:179)
   by 0x417778: ped_unit_format_custom_byte (unit.c:220)
   by 0x417A90: ped_unit_format (unit.c:297)
   by 0x40CD85: do_print (parted.c:1517)
   by 0x40A035: command_run (command.c:139)
   by 0x41210C: non_interactive_mode (ui.c:1540)
   by 0x40EE41: main (parted.c:2482)

10 bytes in 4 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416E9B: ped_malloc (libparted.c:234)
   by 0x417553: ped_strdup (unit.c:179)
   by 0x4177AC: ped_unit_format_custom_byte (unit.c:220)
   by 0x417AC4: ped_unit_format (unit.c:297)
   by 0x40CE34: do_print (parted.c:1527)
   by 0x40A035: command_run (command.c:139)
   by 0x412140: non_interactive_mode (ui.c:1540)
   by 0x40EE75: main (parted.c:2485)

16 years agoplug two more leaks in gpt_write
Jim Meyering [Thu, 29 May 2008 19:48:14 +0000 (21:48 +0200)]
plug two more leaks in gpt_write

* libparted/labels/gpt.c (gpt_write):
1,024 bytes in 2 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416F1B: ped_malloc (libparted.c:270)
   by 0x43E967: pth_get_raw (gpt.c:334)
   by 0x44039D: gpt_write (gpt.c:1067)
   by 0x418F9A: ped_disk_commit_to_dev (disk.c:486)
   by 0x418FE0: ped_disk_commit (disk.c:509)
   by 0x40AF7A: do_mklabel (parted.c:622)
   by 0x40A055: command_run (command.c:139)
   by 0x4121A4: non_interactive_mode (ui.c:1540)

16 years agoplug a blatant leak in gpt_write
Jim Meyering [Thu, 29 May 2008 18:53:08 +0000 (20:53 +0200)]
plug a blatant leak in gpt_write

* libparted/labels/gpt.c (gpt_write):
1,536 bytes in 3 blocks are definitely lost in loss record 9 of 11
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416F1B: ped_malloc (libparted.c:270)
   by 0x44021F: gpt_write (gpt.c:1036)
   by 0x418F9A: ped_disk_commit_to_dev (disk.c:486)
   by 0x418FE0: ped_disk_commit (disk.c:509)
   by 0x40AF7A: do_mklabel (parted.c:622)
   by 0x40A055: command_run (command.c:139)
   by 0x4121A4: non_interactive_mode (ui.c:1540)
   by 0x40EED8: main (parted.c:2499)

16 years agoplug leaks in gpt_write
Jim Meyering [Thu, 29 May 2008 18:47:44 +0000 (20:47 +0200)]
plug leaks in gpt_write

* libparted/labels/gpt.c (gpt_write):
6,674 bytes in 11 blocks are definitely lost...
  at 0x4A0739E: malloc (vg_replace_malloc.c:207)
  by 0x416F1B: ped_malloc (libparted.c:270)
  by 0x43E76D: pth_new (gpt.c:289)
  by 0x43E7E2: pth_new_zeroed (gpt.c:298)
  by 0x43FF61: _generate_header (gpt.c:970)
  by 0x440434: gpt_write (gpt.c:1074)
  by 0x418F9A: ped_disk_commit_to_dev (disk.c:486)
  by 0x418FE0: ped_disk_commit (disk.c:509)
  by 0x40BB35: do_mkpartfs (parted.c:990)

16 years agoplug a leak in gpt_read
Jim Meyering [Thu, 29 May 2008 18:38:13 +0000 (20:38 +0200)]
plug a leak in gpt_read

* libparted/labels/gpt.c (gpt_read):
7,195 bytes in 13 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416F1B: ped_malloc (libparted.c:270)
   by 0x43E759: pth_new (gpt.c:287)
   by 0x43E83C: pth_new_from_raw (gpt.c:310)
   by 0x43F329: _read_header (gpt.c:627)
   by 0x43FB31: gpt_read (gpt.c:826)
   by 0x41882B: ped_disk_new (disk.c:210)
   by 0x40B773: do_mkpartfs (parted.c:884)
   by 0x40A055: command_run (command.c:139)

16 years agoplug leak in ped_device_get_constraint
Jim Meyering [Thu, 29 May 2008 17:38:09 +0000 (19:38 +0200)]
plug leak in ped_device_get_constraint

496 (208 direct, 288 indirect) bytes in 5 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416EDF: ped_malloc (libparted.c:270)
   by 0x41CB40: ped_geometry_new (geom.c:79)
   by 0x415167: ped_device_get_constraint (device.c:438)
   by 0x40B928: do_mkpartfs (parted.c:927)
   by 0x40A055: command_run (command.c:139)
   by 0x41217C: non_interactive_mode (ui.c:1540)
   by 0x40EEB1: main (parted.c:2497)

16 years agoplug leaks in parted.c
Jim Meyering [Thu, 29 May 2008 17:36:02 +0000 (19:36 +0200)]
plug leaks in parted.c

432 (144 direct, 288 indirect) bytes in 3 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416EF3: ped_malloc (libparted.c:270)
   by 0x41D9D2: ped_constraint_new (constraint.c:100)
   by 0x4151A4: ped_device_get_constraint (device.c:439)
   by 0x40B928: do_mkpartfs (parted.c:927)
   by 0x40A055: command_run (command.c:139)
   by 0x41217C: non_interactive_mode (ui.c:1540)
   by 0x40EEB1: main (parted.c:2497)

And another:

* parted/parted.c (do_mkpartfs): Here's one of the signatures:
288 (96 direct, 192 indirect) bytes in 2 blocks are definitely lost...
   at 0x4A0739E: malloc (vg_replace_malloc.c:207)
   by 0x416EF3: ped_malloc (libparted.c:270)
   by 0x41D9D2: ped_constraint_new (constraint.c:100)
   by 0x41DE2C: ped_constraint_intersect (constraint.c:232)
   by 0x40B971: do_mkpartfs (parted.c:930)
   by 0x40A055: command_run (command.c:139)
   by 0x41217C: non_interactive_mode (ui.c:1540)
   by 0x40EEB1: main (parted.c:2497)

* parted/parted.c (do_mkpart): Likewise for this function,
since do_mkpart is nearly identical to do_mkpartfs.

16 years agoplug a leak in ped_device_get_constraint
Jim Meyering [Thu, 29 May 2008 13:47:21 +0000 (15:47 +0200)]
plug a leak in ped_device_get_constraint

* libparted/device.c (ped_device_get_constraint):
512 (224 direct, 288 indirect) bytes in 6 blocks are definitely lost...
  malloc (vg_replace_malloc.c:207)
  ped_malloc (libparted.c:270)
  ped_alignment_new (natmath.c:153)
  ped_device_get_constraint (device.c:432)
  do_mkpartfs (parted.c:927)
  command_run (command.c:139)
  non_interactive_mode (ui.c:1540)
  main (parted.c:2497)

16 years agoAvoid "comparison between signed and unsigned" warning from gcc.
Colin Watson [Thu, 29 May 2008 20:27:13 +0000 (22:27 +0200)]
Avoid "comparison between signed and unsigned" warning from gcc.

* libparted/fs/fat/traverse.c (fat_dir_entry_get_name): Change
type of index from int to size_t.

16 years agoremove more useless "if" tests before free
Jim Meyering [Sun, 4 May 2008 07:55:00 +0000 (09:55 +0200)]
remove more useless "if" tests before free

* libparted/arch/gnu.c (gnu_sync):
* libparted/cs/natmath.c (ped_alignment_destroy):
* libparted/exception.c (ped_exception_catch):
* libparted/labels/dos.c (msdos_partition_destroy):
* parted/parted.c (do_mkpart, do_mkpartfs):

16 years agoUse free, not ped_free.
Jim Meyering [Sun, 4 May 2008 07:51:33 +0000 (09:51 +0200)]
Use free, not ped_free.

* libparted/libparted.c (ped_free): Remove function.
[DEBUG] (_check_dodgy_pointer): Remove function.
Global substitution:
perl -pi -e 's/\bped_free\b/free/g' $(lid -knone ped_free)

16 years ago* maint.mk: Update from coreutils.
Jim Meyering [Thu, 29 May 2008 09:37:51 +0000 (11:37 +0200)]
* maint.mk: Update from coreutils.

16 years agotest for improved DOS-partition-table recognition
Jim Meyering [Thu, 29 May 2008 09:21:05 +0000 (11:21 +0200)]
test for improved DOS-partition-table recognition

* tests/t2200-dos-label-recog.sh: New file.
* tests/Makefile.am (TESTS): Add t2200-dos-label-recog.sh.

16 years agobug fix: improve DOS partition table recognition
Colin Watson [Wed, 28 May 2008 11:12:51 +0000 (12:12 +0100)]
bug fix: improve DOS partition table recognition

* libparted/labels/dos.c (msdos_probe): Make Parted's partition-table
recognition code use the same technique that is used by the Linux
kernel and by util-linux's fdisk.  I.e., accept it whenever all four
boot indicators are 0 or 0x80, rather than using the FAT file system-
recognizing heuristic.
More analysis here:
  http://thread.gmane.org/gmane.comp.gnu.parted.devel/2142/focus=2154
Reported by David Balažic here:
  https://bugs.launchpad.net/ubuntu/+source/parted/+bug/232175
  http://thread.gmane.org/gmane.comp.gnu.parted.devel/2142

16 years agoavoid const-related compiler warnings
Jim Meyering [Fri, 23 May 2008 13:58:20 +0000 (15:58 +0200)]
avoid const-related compiler warnings

* libparted/arch/linux.c (init_dasd, init_generic): Likewise.
* libparted/exception.c (type_strings, option_strings): Likewise.
(ped_exception_get_type_string, ped_exception_get_option_string):
* libparted/fs/ext2/ext2_block_relocator.c (ext2_block_relocator_flush):
* libparted/fs/ext2/ext2_mkfs.c (_set_dirent): Likewise.
* libparted/fs/fat/resize.c (ask_type): Likewise.
* libparted/fs/fat/traverse.c (fat_traverse_begin): Likewise.
* libparted/fs/fat/traverse.h (buffer_size): Likewise.
* libparted/labels/mac.c (_rawpart_cmp_type, _rawpart_cmp_name):
(_rawpart_is_partition_map, _rawpart_is_boot, _rawpart_is_driver):
(_rawpart_has_driver): Likewise.
* parted/parted.c (options, options_help, number_msg): Likewise.
(label_type_msg_start, flag_msg_start, unit_msg_start): Likewise.
(part_type_msg, fs_type_msg_start, start_end_msg, state_msg): Likewise.
(device_msg, name_msg, resize_msg_start, copyright_msg, do_print):
* parted/ui.c (prog_name, banner_msg, usage_msg, bug_msg): Likewise.
(screen_width): Likewise.
* parted/ui.h (prog_name): Likewise.

16 years agoproperly test for __s390__ and __s390x__ macros to disable O_DIRECT on that
Otavio Salvador [Tue, 6 May 2008 17:09:37 +0000 (14:09 -0300)]
properly test for __s390__ and __s390x__ macros to disable O_DIRECT on that

16 years agoRemove useless "if" tests before free.
Jim Meyering [Wed, 30 Apr 2008 16:35:44 +0000 (18:35 +0200)]
Remove useless "if" tests before free.

* libparted/arch/linux.c (_dm_add_partition): Remove useless test.
* libparted/labels/fdasd.c (fdasd_cleanup): Likewise.
* m4/o-direct.m4 (parted_FIND_USABLE_TEST_DIR): Likewise.
* parted/table.c (table_destroy): Likewise.

16 years agoPass NEWS-checking part of "make syntax-check".
Jim Meyering [Tue, 29 Apr 2008 06:47:24 +0000 (08:47 +0200)]
Pass NEWS-checking part of "make syntax-check".

* NEWS: Tweak format so it passes.
* cfg.mk (old_NEWS_hash): Update MD5 checksum.

16 years agoExempt a build-related script from the copyright check.
Jim Meyering [Tue, 29 Apr 2008 06:42:22 +0000 (08:42 +0200)]
Exempt a build-related script from the copyright check.

* .x-sc_GPL_version: New file.
* Makefile.am (EXTRA_DIST): Add .x-sc_GPL_version.

16 years agoTurn off strcmp/STREQ check.
Jim Meyering [Tue, 29 Apr 2008 06:39:46 +0000 (08:39 +0200)]
Turn off strcmp/STREQ check.

* cfg.mk (local-checks-to-skip): Add sc_prohibit_strcmp.

16 years ago* bootstrap.conf (gnulib_modules): Add useless-if-before-free.
Jim Meyering [Tue, 29 Apr 2008 06:35:48 +0000 (08:35 +0200)]
* bootstrap.conf (gnulib_modules): Add useless-if-before-free.

16 years ago* doc/parted.texi: Update copyright dates.
Jim Meyering [Tue, 29 Apr 2008 06:34:52 +0000 (08:34 +0200)]
* doc/parted.texi: Update copyright dates.

16 years agomore updates from gnulib and coreutils
Jim Meyering [Tue, 29 Apr 2008 06:27:22 +0000 (08:27 +0200)]
more updates from gnulib and coreutils

* bootstrap.conf (gnulib_modules): Add gitlog-to-changelog.
* .gitignore: Add .tarball-version
* Makefile.am (EXTRA_DIST): Add .version and .prev-version.
(dist-hook): Update rule.
(gen-ChangeLog): New rule.
* maint.mk: Rename from Makefile.maint and update from coreutils.
* cfg.mk: Rename from Makefile.cfg and update from coreutils.

16 years agouse gnulib's gnumakefile module
Jim Meyering [Tue, 29 Apr 2008 06:08:32 +0000 (08:08 +0200)]
use gnulib's gnumakefile module

* GNUmakefile: Remove from version control.
* bootstrap.conf (gnulib_modules): Add gnumakefile.
* .gitignore: Add GNUmakefile

Signed-off-by: Jim Meyering <meyering@redhat.com>
16 years agoCorrected a few memory leaks from unallocated ped_unit_format calls.
Matt Davis [Tue, 29 Apr 2008 02:56:59 +0000 (22:56 -0400)]
Corrected a few memory leaks from unallocated ped_unit_format calls.

Deallocated memory in do_print() allocated by ped_unit_format calls.

16 years agoCorrected memory leak when displaying partition flags.
Matt Davis [Tue, 29 Apr 2008 02:28:49 +0000 (22:28 -0400)]
Corrected memory leak when displaying partition flags.

Deallocated memory allocated by partition_print_flags()

16 years agoCorrected a small memory leak when displaying partition information
Matt Davis [Sun, 27 Apr 2008 00:31:39 +0000 (20:31 -0400)]
Corrected a small memory leak when displaying partition information

Deallocated a very small memory allocation that occurs in a for loop.
This occurred during output of partition numbers.

16 years agodo not loop in case message exception has no message
Otavio Salvador [Thu, 24 Apr 2008 20:46:28 +0000 (17:46 -0300)]
do not loop in case message exception has no message

The exception handler was looping when the exception had no message.

16 years agofix memory leak in exception handler
Otavio Salvador [Thu, 24 Apr 2008 18:55:47 +0000 (15:55 -0300)]
fix memory leak in exception handler

The exception handler wasn't freeing the memory when the allocated
space wasn't enough to store the error message thus leaving unmanaged
memory around.

16 years agoAdd support for GRUB / BIOS partition in GPT.
Robert Millan [Mon, 25 Feb 2008 20:51:13 +0000 (21:51 +0100)]
Add support for GRUB / BIOS partition in GPT.

16 years agoFix syntax error.
Bastian Blank [Mon, 7 Apr 2008 11:18:45 +0000 (13:18 +0200)]
Fix syntax error.

16 years agoUse host_cpu for s390 check.
Bastian Blank [Mon, 7 Apr 2008 11:07:12 +0000 (13:07 +0200)]
Use host_cpu for s390 check.

16 years agoFix a typo: s/to defragmenting/to defragment/, Remove trailing blanks.
Jim Meyering [Sun, 30 Mar 2008 12:15:26 +0000 (14:15 +0200)]
Fix a typo: s/to defragmenting/to defragment/, Remove trailing blanks.

16 years agoDon't write into line[-1] when line starts with a NUL byte.
Jim Meyering [Mon, 4 Feb 2008 07:35:17 +0000 (08:35 +0100)]
Don't write into line[-1] when line starts with a NUL byte.

* parted/ui.c (_readline): Check strlen first.

16 years agodvh.c (dvh_partition_set_name): Include partition name in diagnostic.
Jim Meyering [Sat, 2 Feb 2008 16:00:42 +0000 (17:00 +0100)]
dvh.c (dvh_partition_set_name): Include partition name in diagnostic.

16 years agoEnforce inherent limitations of dos and dvh partition table formats.
Jim Meyering [Thu, 10 Jan 2008 13:51:56 +0000 (14:51 +0100)]
Enforce inherent limitations of dos and dvh partition table formats.

* libparted/disk.c (_check_partition): Enforce the 32-bit limitation
on a partition's starting sector number and length (in sectors).
With the usual 512-byte sector size, this limits the maximum
partition size to just under 2TB.
(_partition_max_start, _partition_max_len): New functions.
(_check_partition): Use them.
* tests/t4100-msdos-partition-limits.sh: New file.  Test vs. msdos.
* tests/t4100-dvh-partition-limits.sh: New file.  Test vs. dvh.
* tests/Makefile.am (TESTS): Add t4100-msdos-partition-limits.sh
and t4100-dvh-partition-limits.sh.

16 years agomkpart: Don't require a DVH partition name if it's guaranteed to fail.
Jim Meyering [Sat, 2 Feb 2008 19:57:01 +0000 (20:57 +0100)]
mkpart: Don't require a DVH partition name if it's guaranteed to fail.

The mkpart command has an undocumented feature whereby it prompts for
(interactive) or requires (-s) a partition name, *regardless* of whether
it already knows the partition type (any thing but 'logical') is
incompatible with a name.

At first I was pissed and simply #if-0'd the offending code.
But in case someone is actually relying on it, I've relented, and
merely remove the prompt/requirement when the partition table type
is "dvh" and the type of the partition in question is not "logical".

* parted/parted.c (do_mkpart):

16 years agoAvoid new error detected by very latest gcc.
Jim Meyering [Mon, 14 Jan 2008 19:01:39 +0000 (20:01 +0100)]
Avoid new error detected by very latest gcc.

* libparted/fs/fat/traverse.c (fat_dir_entry_get_name): Don't reference
->extension[3] via a pointer into the prior ->name[8] struct member.
gcc detected the reference beyond end of name[8].
Declare first parameter to be "const".
* libparted/fs/fat/traverse.c: Update prototype.

16 years ago#ifdef function definitions to match uses, to avoid compiler warnings.
Jim Meyering [Mon, 14 Jan 2008 17:50:54 +0000 (18:50 +0100)]
#ifdef function definitions to match uses, to avoid compiler warnings.

* parted/ui.c (mask_signal, s_sigint_handler, s_sigsegv_handler)
(s_sigfpe_handler):

Signed-off-by: Jim Meyering <meyering@redhat.com>
16 years agoChange two "extern inline" functions to "static inline".
Jim Meyering [Mon, 14 Jan 2008 17:48:38 +0000 (18:48 +0100)]
Change two "extern inline" functions to "static inline".

* include/parted/natmath.h (ped_div_round_up): This makes
it compilable with bleeding-edge gcc.
(ped_div_round_to_nearest): Likewise.
* libparted/cs/natmath.c (ped_div_round_up, ped_div_round_to_nearest):
Remove definitions.

Signed-off-by: Jim Meyering <meyering@redhat.com>
16 years agoMake inter-release --version output more useful.
Jim Meyering [Mon, 17 Dec 2007 20:44:43 +0000 (21:44 +0100)]
Make inter-release --version output more useful.

Now, each unofficial build has a version "number" like 1.8.8.1.19-58dd,
which indicates that it is built using the 19th change set
(in _some_ repository) following the "v1.8.8.1" tag, and that 58dd
is a prefix of the commit SHA1.
* configure.ac: Run it to set the version.
(PED_MAJOR_VERSION): Derive from $PACKAGE_VERSION.
(PED_MINOR_VERSION): Likewise.
(PED_MICRO_VERSION): Likewise.
Remove the test that would ensure $PACKAGE_VERSION != $PED_VERSION,
now that the latter is derived from the former.
* Makefile.am (dist-hook): Arrange so that .version appears only
in distribution tarballs, never in a checked-out repository.
* .gitignore: Add .version here, too.  Just in case.
* lib/.gitignore: Remove now-generated (by bootstrap) file.
* build-aux/.gitignore: Likewise.

Ensure that $(VERSION) is up to date for dist-related targets.
* GNUmakefile: Arrange to rerun autoconf, if the version reported by
git-version-gen doesn't match $(VERSION), but only for dist targets.

16 years agodoc/pt_BR/partprobe.8.pt_BR.po: Remove trailing space in generated file.
Jim Meyering [Tue, 18 Dec 2007 08:38:41 +0000 (09:38 +0100)]
doc/pt_BR/partprobe.8.pt_BR.po: Remove trailing space in generated file.

16 years agoDistribute new file: architecture.h.
Jim Meyering [Tue, 18 Dec 2007 08:24:52 +0000 (09:24 +0100)]
Distribute new file: architecture.h.

libparted/Makefile.am (libparted_la_SOURCES): Add architecture.h.

16 years agolibparted/architecture.c: Include <config.h>.
Jim Meyering [Tue, 18 Dec 2007 08:22:34 +0000 (09:22 +0100)]
libparted/architecture.c: Include <config.h>.

16 years agoFix up the ChangeLog generation a bit.
David Cantrell [Mon, 17 Dec 2007 22:04:06 +0000 (12:04 -1000)]
Fix up the ChangeLog generation a bit.

We can just take the output of git-log.

16 years agoPerform ChangeLog-creation here, rather than in bootstrap,
Jim Meyering [Mon, 17 Dec 2007 20:53:58 +0000 (21:53 +0100)]
Perform ChangeLog-creation here, rather than in bootstrap,

so that the next merge will be less likely to clobber this parted-specific bit.

16 years agoMention that we should be able to build with -fno-common.
Jim Meyering [Mon, 17 Dec 2007 15:50:23 +0000 (16:50 +0100)]
Mention that we should be able to build with -fno-common.

16 years ago* bootstrap: Update from Coreutils, adapting to changes in gnulib.
Jim Meyering [Mon, 17 Dec 2007 15:42:42 +0000 (16:42 +0100)]
* bootstrap: Update from Coreutils, adapting to changes in gnulib.

Among them: now gnulib uses git for version control.

16 years ago* libparted/tests/disk.c (START_TEST): Remove decl of unused local.
Jim Meyering [Mon, 17 Dec 2007 15:37:08 +0000 (16:37 +0100)]
* libparted/tests/disk.c (START_TEST): Remove decl of unused local.

16 years agoAdd flags for BSD disklabels (needed for Linux on Alpha)
David Cantrell [Wed, 7 Nov 2007 16:52:47 +0000 (11:52 -0500)]
Add flags for BSD disklabels (needed for Linux on Alpha)

Patch from Sergey Tikhonov from the AlphaCore project.

16 years agoAdd KNOWN ISSUES section
David Cantrell [Tue, 6 Nov 2007 20:13:54 +0000 (15:13 -0500)]
Add KNOWN ISSUES section

Add a KNOWN ISSUES section explaining that ext3 resizing does not work
and that you should use resize2fs.

16 years agoAlways define PED_DEVICE_DM
David Cantrell [Tue, 6 Nov 2007 20:13:15 +0000 (15:13 -0500)]
Always define PED_DEVICE_DM

Regardless of compile time options, always define PED_DEVICE_DM.

16 years agoAdd 'xvd' to the transports array in do_print
David Cantrell [Tue, 6 Nov 2007 20:12:29 +0000 (15:12 -0500)]
Add 'xvd' to the transports array in do_print

For the parted print command, add 'xvd' to the transports array so there is
a matching string to display when we print a line for Xen virtual block
device.

16 years agoMove PedArchitecture and ped_set_architecture to private
Otavio Salvador [Fri, 2 Nov 2007 16:33:05 +0000 (14:33 -0200)]
Move PedArchitecture and ped_set_architecture to private

Code that works with libparted isn't suppose to need to change the
architecture where it's running and then doesn't make sense to export
it.

Two new files has been create (libparted/architecture.[hc]) that has
the PedArchitecture structure and the ped_set_architecture
implementation.  All changes that were need to get it suported on all
currently available architectures has been done too.

16 years agoparted/geom.h: move includes to after type definition to avoids dependency
Otavio Salvador [Fri, 2 Nov 2007 20:42:52 +0000 (18:42 -0200)]
parted/geom.h: move includes to after type definition to avoids dependency

16 years agoparted/device.h: move includes to after type definition to avoids dependency
Otavio Salvador [Fri, 2 Nov 2007 20:42:48 +0000 (18:42 -0200)]
parted/device.h: move includes to after type definition to avoids dependency

16 years agoparted/filesys.h doesn't use stdio.h, parted/disk.h and parted/exception.h
Otavio Salvador [Fri, 2 Nov 2007 16:57:32 +0000 (14:57 -0200)]
parted/filesys.h doesn't use stdio.h, parted/disk.h and parted/exception.h

16 years agoparted/constraint.h needs parted/geom.h since it uses PedGeometry
Otavio Salvador [Fri, 2 Nov 2007 14:22:03 +0000 (12:22 -0200)]
parted/constraint.h needs parted/geom.h since it uses PedGeometry

16 years agoparted/natmath.h needs parted/device.h since it uses PedSector
Otavio Salvador [Fri, 2 Nov 2007 14:15:22 +0000 (12:15 -0200)]
parted/natmath.h needs parted/device.h since it uses PedSector

16 years agolibparted/cs/natmath.c: remove inline since it's not supported by GNU99 standard
Otavio Salvador [Fri, 2 Nov 2007 14:14:36 +0000 (12:14 -0200)]
libparted/cs/natmath.c: remove inline since it's not supported by GNU99 standard

16 years agoparted/device.h needs to include parted/constraint.h since it uses PedConstraint
Otavio Salvador [Fri, 2 Nov 2007 13:54:42 +0000 (11:54 -0200)]
parted/device.h needs to include parted/constraint.h since it uses PedConstraint

16 years agoRemoving unnecessary type-casts and eerie comments.
Debarshi Ray [Fri, 2 Nov 2007 11:53:32 +0000 (17:23 +0530)]
Removing unnecessary type-casts and eerie comments.

16 years agoAdded information to message about ext2 incompatible features.
Mark Neyhart [Mon, 29 Oct 2007 18:15:03 +0000 (10:15 -0800)]
Added information to message about ext2 incompatible features.

Modifed the message about incompatible ext2 features installed to include a
list of compatible features and to recommend the usage of tune2fs or
debugfs to remove features which are not compatible.

16 years agoReally duplicate the disk instead of readd every partition
Otavio Salvador [Thu, 30 Aug 2007 12:52:03 +0000 (09:52 -0300)]
Really duplicate the disk instead of readd every partition

To avoid possible differences between the original disk layout and the
duplicated one, a raw copy is done. Has been identified a case[1]
where extended partitions had their positions changed due this.

 1. http://bugs.debian.org/294520

The recipe[2] to reproduce the problem, on the provided URI, has been
used to produced a test and hence be sure it's not forgotten anymore.

 2. http://bugs.debian.org/294520#34

The fix has been produced by Samuel Thibault <samuel.thibault@ens-lyon.org>

17 years agoSet partprobe's text domain.
Benno Schulenberg [Thu, 16 Aug 2007 14:24:50 +0000 (16:24 +0200)]
Set partprobe's text domain.

* partprobe/partprobe.c (main): Set textdomain.

17 years agoFix mkpart linux-swap bug: would use 0x83 rather than 0x82
Jim Meyering [Mon, 13 Aug 2007 20:36:27 +0000 (22:36 +0200)]
Fix mkpart linux-swap bug: would use 0x83 rather than 0x82

* libparted/labels/bsd.c (bsd_partition_set_system): Include "misc.h".
Use is_linux_swap to test whether the type string matches.
* libparted/labels/dasd.c (dasd_read, dasd_partition_set_system): Likewise.
* libparted/labels/dos.c (msdos_partition_set_system): Likewise.
* libparted/labels/mac.c (mac_partition_set_system): Likewise.
* libparted/labels/rdb.c (amiga_partition_set_system): Likewise.
* libparted/labels/sun.c (sun_partition_set_system): Likewise.
Based on a patch by Kenneth MacDonald, from
<http://lists.gnu.org/archive/html/bug-parted/2007-07/msg00012.html>.
* libparted/labels/misc.h (is_linux_swap): New function/file.
* libparted/labels/Makefile.am (liblabels_la_SOURCES): Add misc.h.
* tests/t2100-mkswap.sh: New file, test for the above fix.
* tests/Makefile.am (TESTS): Add t2100-mkswap.sh.

17 years agoDon't try to avoid "free (x)" when x is NULL.
Jim Meyering [Mon, 13 Aug 2007 20:41:39 +0000 (22:41 +0200)]
Don't try to avoid "free (x)" when x is NULL.

* libparted/arch/linux.c (linux_read):
Part of http://git.debian.org/?p=parted/parted.git;a=commitdiff;h=080d3e7078
changed this:

  free(diobuf);

to this:

  if (diobuf)
      free(diobuf);

Yet, free doesn't have any problem with NULL.
Some very old (non-POSIX) implementations did (like SunOS4),
but even for them, gnulib's lib/free.c protects us.

Signed-off-by: Jim Meyering <jim@meyering.net>
17 years agoAvoid test failure with dash's builtin printf.
Jim Meyering [Tue, 14 Aug 2007 09:06:29 +0000 (11:06 +0200)]
Avoid test failure with dash's builtin printf.
* tests/t0100-print.sh (msdos_magic): Use more-portable octal escapes,
not hexadecimal ones.

17 years agoSanity check for dev, safety checks on diobuf.
David Cantrell [Wed, 8 Aug 2007 22:04:00 +0000 (18:04 -0400)]
Sanity check for dev, safety checks on diobuf.

Make sure diobuf is NULL before we begin and make sure it isn't NULL when
we try to free it.  Throw an exception if dev is NULL and we enter this
function.
(cherry picked from commit be2ace6e8d381fb836647234ac65d34a4a547e31)

17 years agoRevert "History with undo and redo capabilities."
David Cantrell [Thu, 9 Aug 2007 15:20:44 +0000 (11:20 -0400)]
Revert "History with undo and redo capabilities."

This reverts commit 3bb8494e1ed5af0a48ad0211c3219e653167854f.

17 years agoHistory with undo and redo capabilities.
David Cantrell [Wed, 8 Aug 2007 22:14:08 +0000 (18:14 -0400)]
History with undo and redo capabilities.

Author: Matt Davis <mattdavis@gmail.com>

Here is a working version of the history with undo/redo capabilities.  The
idea here was based on a talk with Otavio Salvador who mentioned the concepts
from Vanni Brutto.  The idea being to capture all changes and only apply them
all at once.  This protects the user from performing an unwanted change.  What
I did was capture all disk modifications that are committed to disk, and put
them in a list.  The history manager allows the list to be traversed linearly,
so that a change can be undone, viewed in parted (print command), or reapplied
again.  Nothing actually happens to the physical disk until the 'save' command
is issued.

Jim Meyering suggested that the functionality might be useful to libparted
thus most of the functionality has been moved there.  The stdout displays
(printf) are placed in parted.c as the library should not do any output
printing on its own.

Three commands were added:
1) Undo : Undoes a disk modification
2) Redo : Redoes the most recent 'undone' modification
3) Save : Actually commits the list of non-undone modifications to disk

I feel a bit more testing needs to be done, but I am happy with the results
right now.

17 years agoRevert "If we cannot create a new PedDisk for the device we're looking at, return...
Jim Meyering [Wed, 1 Aug 2007 07:44:56 +0000 (09:44 +0200)]
Revert "If we cannot create a new PedDisk for the device we're looking at, return NULL rather than the device path."
The reverted change causes almost all tests to fail.

This reverts commit a6f86144c0b5fd07c462e256bc7b54095c7b8e7a.

17 years agoRevert "Removed unused label (compiler warning)."
Jim Meyering [Wed, 1 Aug 2007 07:37:13 +0000 (09:37 +0200)]
Revert "Removed unused label (compiler warning)."
The label *is* used.

This reverts commit f4bc90ac22de605ad927d2d85ba468a16bf55ae7.

17 years agoProbe for all device-mapper devices.
David Cantrell [Tue, 31 Jul 2007 19:41:22 +0000 (15:41 -0400)]
Probe for all device-mapper devices.

Add _dm_probe_all() from Debian and patch linux_probe_all() to probe for
device-mapper devices after probing for standard devices.
(cherry picked from commit 609b7ae6d274e479027bb46c4bd10227cd921803)

17 years agoAdd the PED_DEVICE_XVD device type for Xen virtual block devices.
David Cantrell [Tue, 31 Jul 2007 17:57:55 +0000 (13:57 -0400)]
Add the PED_DEVICE_XVD device type for Xen virtual block devices.
(cherry picked from commit af4bea2ac854343609f3e6688bedd727ddf76ae6)

17 years agoDetect Xen virtual block devices and identify them as such.
David Cantrell [Tue, 31 Jul 2007 17:58:40 +0000 (13:58 -0400)]
Detect Xen virtual block devices and identify them as such.
(cherry picked from commit 64861efcaae95f605614d5e4c6d9c35c5d5712d7)

17 years agoIf we cannot create a new PedDisk for the device we're looking at, return NULL rather...
David Cantrell [Tue, 31 Jul 2007 17:59:27 +0000 (13:59 -0400)]
If we cannot create a new PedDisk for the device we're looking at, return NULL rather than the device path.
(cherry picked from commit 865ea3d3f2eb7918b64a00825dfa44e05651e2ad)

17 years agoWhen reading the DASD disk label, look at the partition flags as well as what is...
David Cantrell [Tue, 31 Jul 2007 18:00:31 +0000 (14:00 -0400)]
When reading the DASD disk label, look at the partition flags as well as what is on the actual partition using ped_file_system_probe().  This avoids flags being set for partitions when they shouldn't be.
(cherry picked from commit dfafc45a3775cc92d3cd89c9b35926cab23db512)

17 years agoRemoved unused label (compiler warning).
David Cantrell [Tue, 31 Jul 2007 18:06:09 +0000 (14:06 -0400)]
Removed unused label (compiler warning).
(cherry picked from commit b43bda3134059cdacb13e1a4d0a0bee8a543cfb0)

17 years agoFix bootstrap script to support the new translationproject html files
Otavio Salvador [Fri, 27 Jul 2007 01:46:31 +0000 (22:46 -0300)]
Fix bootstrap script to support the new translationproject html files

17 years agoUpdated GPL license boilerplate.
David Cantrell [Mon, 23 Jul 2007 17:54:23 +0000 (13:54 -0400)]
Updated GPL license boilerplate.

17 years agoUpdated COPYING file to the GNU General Public License version 3. Updated boilerplat...
David Cantrell [Mon, 23 Jul 2007 17:07:02 +0000 (13:07 -0400)]
Updated COPYING file to the GNU General Public License version 3.  Updated boilerplate GPL text to reference version 3 of the license.

17 years agoFlip the switch (expected-failure -> expected-success) in tests/t3100-
Jim Meyering [Tue, 10 Jul 2007 10:03:49 +0000 (12:03 +0200)]
Flip the switch (expected-failure -> expected-success) in tests/t3100-
resize-ext2-partion.sh, now that this bug is fixed.

17 years agoFix block state checking for realocated blocks
Flavio Leitner [Tue, 12 Jun 2007 17:54:26 +0000 (14:54 -0300)]
Fix block state checking for realocated blocks

A busy block should be realocated and it's correct in
ext2_block_relocator_mark(), but not in ext2_metadata_push().

Signed-off-by: Flavio Leitner <flavio.leitner@gmail.com>
Signed-off-by: Jim Meyering <jim@meyering.net>
17 years agoFix block number used when checking for state
Flavio Leitner [Tue, 12 Jun 2007 17:53:20 +0000 (14:53 -0300)]
Fix block number used when checking for state

Hi there,

The ext2_bread() returns a descriptor containing a
pointer ->data representing the contents of 1 block.

In ext2_block_relocate_grow(), it reads the block bitmap from
a group descriptor representing a range of blocks:
   bh = ext2_bread(fs, EXT2_GROUP_BLOCK_BITMAP(fs->gd[i]));

Then it does:
   k = EXT2_GROUP_INODE_TABLE(fs->gd[i]) + fs->inodeblocks + j;
k is the absolute block number and then checks the state doing:
if (bh->data[k>>3] & _bitmap[k&7])

The k should be the offset inside of group descriptor and not
the absolute block number. Example:
. Block bitmap represents 512 blocks
. Block absolute number is 1023.

GrpDesc = Block absolute number / block bitmap size = 1
bh = ext2_bread(fs, EXT2_GROUP_BLOCK_BITMAP(fs->gd[GrpDesc]))
bh->data[] contains a bitmap of 512 blocks from 512-1024
relative = absolute block number % block bitmap size
relative = 1023/512 = 511

The block state is in bitmap bh->data[relative>>3] & ...

Signed-off-by: Flavio Leitner <flavio.leitner@gmail.com>
Signed-off-by: Jim Meyering <jim@meyering.net>
17 years agoFix syntax error and LIBS problem in configure.ac
Michael Brennan [Tue, 10 Jul 2007 02:26:38 +0000 (04:26 +0200)]
Fix syntax error and LIBS problem in configure.ac

Make sure LIBS doesn't contain libraries that will prevent
successful compilations after libreadline has been tested.
A trailing comma generated a syntax error in the configure script.

Signed-off-by: Michael Brennan <brennan.brisad@gmail.com>
17 years agoMore correct handling of the HeaderSize field in GPT labels
Matthew S. Harris [Sun, 8 Jul 2007 19:27:07 +0000 (12:27 -0700)]
More correct handling of the HeaderSize field in GPT labels

- Use the HeaderSize field value when determining how many bytes to
compute the CRC over.

- Don't abort if the HeaderSize field value is bigger than our struct,
since more fields may be defined in the future.

Signed-off-by: Matthew S. Harris <mharris@coruscant.(none)>
17 years agoDon't let the automatically-generated ChangeLog cause "make dist" failure.
Jim Meyering [Fri, 6 Jul 2007 19:29:46 +0000 (21:29 +0200)]
Don't let the automatically-generated ChangeLog cause "make dist" failure.
* Makefile.cfg (local-checks-to-skip): Add sc_changelog.

17 years agoLet "make distcheck" pass once again.
Jim Meyering [Fri, 6 Jul 2007 11:53:39 +0000 (13:53 +0200)]
Let "make distcheck" pass once again.
* parted/strlist.c (str_list_alloc): Remove unnecessary
cast of xmalloc return value.

17 years agoFix a problem where a partial read would not be handled correctly;
Michael Brennan [Fri, 6 Jul 2007 11:39:39 +0000 (13:39 +0200)]
Fix a problem where a partial read would not be handled correctly;
also change the status variable to ssize_t.

17 years agoDon't include config.h from internal headers.
Jim Meyering [Tue, 3 Jul 2007 08:47:03 +0000 (10:47 +0200)]
Don't include config.h from internal headers.
This avoids redefinition errors (on the new symbol, __STDC_LIMIT_MACROS_TRIGGER)
when config.h is included twice.

Bob Proulx reported the buildbot failures: http://buildbot.proulx.com:9001/

17 years agoAvoid test failure on Linux-2.6.8 due to too small underlying file.
Jim Meyering [Fri, 22 Jun 2007 17:44:27 +0000 (19:44 +0200)]
Avoid test failure on Linux-2.6.8 due to too small underlying file.
Overwrite the label creating a 4KB file; 1KB was too small and would
trigger the failure.  Reported by Bob Proulx.

17 years agoWork around test failures caused by inadequate libreadline.
Jim Meyering [Fri, 22 Jun 2007 14:30:16 +0000 (16:30 +0200)]
Work around test failures caused by inadequate libreadline.

* configure.ac: Reject an inadequate libreadline5.0.
* parted/ui.c (_readline) [!HAVE_LIBREADLINE]: Echo each
just-read line, to be consistent with libreadline5.2.

Signed-off-by: Jim Meyering <jim@meyering.net>
17 years agoAvoid mklabel test failure with very small file on Linux-2.6.8.
Jim Meyering [Fri, 22 Jun 2007 14:01:59 +0000 (16:01 +0200)]
Avoid mklabel test failure with very small file on Linux-2.6.8.

Signed-off-by: Jim Meyering <jim@meyering.net>
17 years agoREADME-hacking: Don't mention Gzip 1.2.4, now that 1.3.12 is out.
Jim Meyering [Mon, 18 Jun 2007 12:52:27 +0000 (14:52 +0200)]
README-hacking: Don't mention Gzip 1.2.4, now that 1.3.12 is out.
Mention uuid-devel and pkg-config, too.

17 years agoReplace some malloc and strdup by xmalloc and xstrdup respectively
David Vazquez [Tue, 12 Jun 2007 19:25:47 +0000 (21:25 +0200)]
Replace some malloc and strdup by xmalloc and xstrdup respectively

I replace some malloc and strdup by xmalloc and xstrdup respectively.

Signed-off-by: David Vazquez <xeos@gmail.com>
17 years agoFix the mkfs final-block-group-too-short bug.
Jim Meyering [Tue, 12 Jun 2007 15:25:18 +0000 (17:25 +0200)]
Fix the mkfs final-block-group-too-short bug.

mkfs would fail for certain sizes resulting in a final block-group
that was too small to accommodate the minimum number of admin blocks.
There was already work-around code for when a decremented "numgroups"
was 1.  This change applies that work-around code for larger values of
numgroups, too.  Also, there was an off-by-one error in the guard test
that would allow a few too-small partition sizes to slip through and
provoke the error (see tests/t2000-mkfs.sh for two examples).
This change fixes that, too.

17 years agoFix two "make check"-as-root failures.
Jim Meyering [Tue, 12 Jun 2007 14:39:55 +0000 (16:39 +0200)]
Fix two "make check"-as-root failures.
t1100-busy-label.sh: Remove extra copy of "o2" temporary file name.
test-lib.sh (emit_superuser_warning): Exit successfully when run as root, too.